Valen Johnson

Since the fall of 2012, I have been a professor in the Department of Statistics at Texas A&M University.  Before this, I was a Professor of Biostatistics at both the University of Texas M.D. Anderson Cancer Center (2004-2012) and the University Michigan (2002-2004).  And before that I was a Professor of Statistics at Duke University (1989-2001) and worked for one year at Los Alamos National Laboratory (2001-2002).   I received my Ph.D. in statistics from The University of Chicago in 1989. My applied research interests include educational assessment, ordinal data and rank data analysis, clinical trial design, image analysis, and reliability analysis.  My current methodological interests focus on Bayesian hypothesis testing and its connections to classical testing procedures, Bayesian variable selection, Markov chain Monte Carlo model diagnostics, and latent variable modeling.  A few recent papers on these topics are listed below.

Selected recent publications:

- Revised Standards for Statistical Evidence (2013), Proceedings of the National Academy of Sciences of the United States of America, 110(48), 19313-19317.
- Uniformly most powerful Bayesian tests (2013), Annals of Statistics, 41(4), 1716-1741
- On numerical aspects of Bayesian variable selection (2013), Bayesian Analysis, 8(4), 741-758
- A Bayesian Dose-Finding Design for Phase I/II Clinical Trials (2014) (software at ReadMe and phase12 )
- Bayesian variable selection in high-dimensional settings (2012), with David Rossell, Journal of the American Statistical Association, 107, 649-660.
On the use of non-local prior densities for default Bayesian hypothesis tests (2010), with David Rossell, Journal of the Royal Statistical Society Series B, 72, 143-170.


Grade Inflation (2003), Springer
Ordinal Data Modeling (1999), with James Albert, Springer-Verlag.

A more complete listing of my publications can be found here

or at Google Scholar